Evan Wimpey: Hello, and welcome to the Mining Your Own Business podcast. I’m your host, Evan Wimpey. And today I’m super excited to introduce our guest, Frazier Seay-Rad. Frazier is an AI and ML leader in the sports betting and casino industry. A very cool field and a very cool industry to be in. Frazier. Excited to hear what you have to say to learn a little bit about the industry and to see what data and analytics are doing for sports betting.
Frazier, thanks for coming on the show.
Frazier Seay-Rad: Yeah. Thanks, Evan. I’m excited to be here. There’s just so much going on. It’s a rapidly developing industry. There’s like tons to work on really cool for analytics, really cool in the data science space. So happy to be here.
Evan Wimpey: Very cool. Thank you so much. And can you tell us a little bit about how you got here? Not here to this podcast episode, but here to this role as a data and analytics leader in. In the sports betting world.
Frazier Seay-Rad: Yeah. I have a really weird journey actually. Not your typical track whatsoever. So I actually thought I wanted to work in the criminal justice system. That was like my plan. I wanted to be an investigator.
My parents didn’t want me to wear a bulletproof vest. So they said, can you do anything else? I ended up, which was literally a conversation we had. So I ended up saying, okay, let me torture some data and figure out like what questions I can ask and all this other stuff. But a little bit before that, I actually worked in audit, which was interesting.
So I always say that’s like corporate police. And I liked it though, because I could ask a lot of questions. I could see a lot. And I, you know, basically ended up Working my way through that and wanting to test more than a normal sample size. Like the audit industry is definitely like a choose 25 samples.
Doesn’t matter the population, but something just doesn’t sit right. Especially if you’re looking at even, you know, like 200 or something or whatever. So I was like, let’s just test the entire population. And I had no idea really what that meant. So I fully went into like nine to five, I’m going to audit.
And then five to nine, I’m going to learn everything I can about what I promised to these very high up people. I was like, no, I’m just going to automate this audit. Like, I’m just going to test the whole population. Like I said, I didn’t know what that meant, which is wild looking back at it. I’m like, who says that?
And I’m pretty sure I was an intern at the time too. I just wanted a full-time position. But I did. We pulled it off. It was successful. We did test all of it. It did take a long time but led to the largest audit findings. And I just found myself like, wow, I feel like I really can do these investigations.
I can ask these questions. I don’t have to wear a bulletproof vest, and my parents are happy. And I did that for a while. I really, really enjoyed it. I was, you know, compliance audit world, is interesting for sure. But then I wanted to expand and work a little bit more on my data skills. So I ended up going and working for a different company with like, instant needs grocery delivery.
And I did that for a while, which I feel like I learned a lot because it was a lot of exposure to different like business areas. So there was like a little bit of the corporate strategy, a little bit of the marketing, little bit of finance, all these different areas and even like operations, which I liked cause it was like rapid fire, all these different spaces.
And then moved over into this, like, casino and sports betting space, started off analytics, focused on casino, didn’t know anything about sports betting like whatsoever. Didn’t really watch sports either, which is pretty interesting. I’ve since become more of a sports fan. I know a thing or two about sports.
Yeah. So now I’m over in the data science space, which is different. I feel like it’s like analytics on steroids a little bit, a little bit of a different sort of skillset, but still really closely related. And I still love my analytics roots. So now I’m here.
Evan Wimpey: Awesome. Yeah. Very, very cool journey. Audit feels like. When you’re trying to scale it, we’re trying to do more. It has parallels with the data science and analytics world we’re in now. Were you leaning on the tools that you’re using today in data science? Were you using those when you were trying to audit or scale up or automate that process? Or how did you sort of come even to the terms of analytics or machine learning?
Frazier Seay-Rad: There was somebody that actually worked in the department who was over what was called data analytics, and he was just such a heavy SQL user, but a lot on the data engine side to just like pulling all this data in. So I learned a lot of that and saw, like, when I tell you what I was looking at was like, PDFs of data one by one.
So if you compare, then it’s like, oh, wait, there’s this whole thing that holds all of this stuff. And mind you, at that time, my experience was just Excel. And barely that, like I could barely merge cells. It’s not like I was in there with these long formulas. Like I would never make it in Excel modeling at the time, but I was like—I didn’t do all this.
So I actually tried to force it into Excel and I was like Excel only has like, you know, 1,100 or I’m sorry, 1.1 million rows or something like that. That’s all you can do. And then your book crashes and you’re screwed and then you can’t send it anywhere anyway. So I found out about Excel’s like power pivot, power query, and this was years ago, right.
And so I was like, wow, I can attach this to a database, do all the Excel things I know and love with the pivot tables. And then somebody is like, why would you even do all that? You can write these simple SQL queries. And I’m like, well, what does that mean? Right. And so then you start looking at, Oh, I have all these, all this data from all these PDFs and I don’t have to look at these PDFs one by one anymore, because if you think about going through just 25 PDFs, it’s like, okay, whatever, like big deal, I can, and manually highlighting too.
I mean, it was, that’s. That’s the name of the audit game. I’ve never thought audit was like really forward thinking when it comes to like innovation, just because it’s just not set up for that. It’s more of like traditional like accounting, you know, not this like super high-tech thing, which I do think is changing, especially as these auditors are having to audit some of these like really crazy systems. So that was basically the start. And then we wanted to do some other projects. That did require a little bit more and I was like, well, I guess my SQL is not going to get me any further. So I had to learn a little Python and that was, I was honestly, I felt way in over my head at first.
I was like, I’m never going to learn. But I just felt like I was looking at like hieroglyphics or something, but I got there and because I don’t have a traditional background, I didn’t take all of the really high-level math. And that didn’t, like, I could only get so far just knowing these languages.
So I went out and bought all the For Dummies books. Because I’m like, nope, that title, alright, that matches. So, we’re talking linear algebra, calculus, you know, like, the works. And I was going through these books. Almost as if you were looking at like an SAT prep book. YouTube videos, Coursera, like all these different sort of things until one day it just kind of clicked a little bit.
But I think it was really fortunate too, because it wasn’t just purely in an academia setting, you know, I was able to see a lot on like hands on, you know, saying like you learn more in the four months on the job after, you know, four years at school. So yeah, it’s interesting for sure.
Evan Wimpey: Very cool.
Yeah, I feel like it’s a lot more motivating to learn linear algebra when you have a real problem that you’re trying to solve where linear algebra is going to help you versus an exam at the end of the semester. That’s great. So. You sort of self-studied on this journey, found the resources that you needed to, to gain the skills that you needed.
Now you’re, you’re leading this, this team of data scientists, engineers, folks that are in this space. Is there anything that you, that you wish other folks, sort of outside of your analytics team knew about analytics or the types of work that you’re doing? Maybe they don’t need to know linear algebra, but is there anything like, boy, if they only knew what we could do or, or the types of skills that we had.
Frazier Seay-Rad: Yeah. Yeah. It’s tough. I feel like it’s two different languages, you know, at the end of the day, you’ve got this very business heavy lingo very much like, what is the value? What is this? What is that? And then you have data science, which is really, in a sense can be a lot of research and the two just clash.
And so how do you bring that together? It definitely takes a lot of, you know, communication and working with these groups. But I wish the one thing I feel like if they just knew more of is like leaning in and not feeling like it’s this black box thing, which historically, I mean, it has been right. If the more explainability that a data scientist can put to what they’re doing and actually turn around and to ask the businessperson, like what business value, what is the problem you’re trying to solve?
Aiming for simplicity first. I mean, it’s getting these two people in a room and then translating between the two until they can figure out how to best communicate. I mean, there’s, cause there’s just so much, but you don’t know what you don’t know. Yeah. And I think a lot of times that’s true for the business side of things is like, I don’t know what’s possible.
I like to get ahead of that. I like to demystify. I like to tell people like, listen, you use data science every day, one way or another, you just don’t realize it. Like. If you like Tik Tok, you got an algorithm. Amazon tells you what items you might want to buy. That’s an algorithm, right. Instagram, same thing.
And so it’s people like, Oh yeah, well that does make sense. And then breaking it down from there. And like I said, just cultivating those relationships and making it seem like you can help them. And I like to always include the data scientists in calls to like, not just myself, but have like a bunch of others in there to weigh in on what the business problem is and start to ask the right questions.
Because I think there’s just this general fear that it’s going to go into this very convoluted, complex math equation. What, and you’re like, nobody wants to look at that. So any way that you can make that portion better, which is why we spend a lot of time saying like, your presentations have to be, you know, very like people have to comprehend it. Well, who is your audience? You know?
Evan Wimpey: Yeah, I think that’s great. Yeah, your audience is not the linear algebra professor from five years ago. It’s somebody who’s in the business trying to make sure they never have to look at linear algebra. Yeah, I think, I think that’s great. And I love a focus on simplicity and trying to do things as simply as, as possible to be effective.
Have you found any challenges there with in the post chat GPT world, where it feels like everything is AI and AI is a solution for everything, which it’s usually not the simplest solution. Have you felt any pressures to get away from that simplicity? Mantra?
Frazier Seay-Rad: No. I mean, I’m a big advocate for AI and I know that a lot of people, like, there’s a lot of mixed opinions.
I think there’s a lot of ways it can help and it’s sometimes it’s easy. It’s like, I want to swap out something in my pasta and I want gluten free. How do I do it? Right. like it, there’s so much you can do. I personally have not had the experience where everyone’s like, I need an AI solution. We actually—we go to a lot of people and say, like, maybe we could leverage like some sort of AI solution, right.
There’s a lot of things that would have taken months or years in the past. So I feel like I almost think about it a little bit of the opposite, but what, what I do have is, or what I think is always interesting, like, if you ask somebody in data engineering, how to like, you need to do something right there and be like, you need the perfect data pipeline.
And you ask somebody who works in just analytics, like you’re going to need the perfect dashboard. And then of course, if you ask a data scientist or like the most complex solution possible, right. I think if you, even if you’re a data scientist, you’re like, this is a five-line SQL code solve. And then you’re generating millions of dollars because of it, then that is a win.
And I think that is the approach that everybody should take first is like, if I had X time to solve this problem, like, how would I do it? Otherwise you’re going to go indefinitely. So a lot of like time boxing. And then if AI is part of the conversation, which I think a lot of times it can be because I’m a big believer.
Like, I think you can be as an organization, you can be first, you can be best, or you can be different, but you can’t be any of those things. If you’re always playing catch up to somebody else. Right. And I feel like AI in so many spaces is that differentiator that could be easy. But if you’re talking about, we need to train on stuff, we need to build this rag, we need to do all of this complex stuff versus like just small deploys of something like Gemini or, you know, whatever, then, then that feels like a lot easier rather than just doing the most complicated thing for the sake of saying you’re doing AI.
Evan Wimpey: Yeah, I think that’s great. I love that sort of first best or different some way to, to have an advantage there. Trying to weigh in my head, if any of those is a priority over the others. That’s a tougher question.
For sure. You’ve got—you lead a team of folks there. You’ve talked about some of the things that are. Important being able to communicate with business folks, trying to build simply bringing them into calls to cross-pollinate across languages there. Can you talk about maybe specifically some of the things that you like to look for in teammates yourself coming from not like a math heavy traditional data science background?
Are there things that you look for when you’re when you’re trying to fill roles?
Frazier Seay-Rad: Yeah, I look for grit, honestly. There’s some Ted talk about that specifies like, I wish I could think of the name, but there’s Angela Duckworth. Yes, exactly. Yes. Like the studies around grit are, I think it’s vital.
Anytime I see somebody thinking outside the box, again, simplicity is a big part of it. But like if you know the math and you know the Python and you know the SQL, right. Like, let’s just say, you know all these things, you can create an end-to-end pipeline. That’s great. So can 10 other people.
But what’s going to make you stand out and make you shine are really your soft skills. Can you put a presentation together? Can you explain this to me? Like I’m five and which is something I commonly tell people, like, explain this to me, like I’m five, because I don’t know everything. Do you have the ability, like if you get knocked down or you have a setback, can you, do you let that bother you and be the end, right.
Some of these projects can span six, seven, eight, nine months. You’re bound to have some sort of hiccup. How do you move forward out of that? Do you pick yourself up and you’re like, I just, I don’t know. I’m going to find a way. Matter of fact, I have on my arm tattooed, literally I’ll find a way or make one.
So I feel like that’s something I definitely want to look for in others. And are you coachable? I feel like that’s a huge one. You know, you see a lot of stuff I think on LinkedIn these days about, yes, you know, the academia and like the. The technical skills are important and they are, and I’m not saying that they’re not, but if you’re in a sea of people that have the exact same things, what is going to make you stand out?
I also look for people that have had some sort of like business courses, especially if they’re just coming out of school. You know, that’s so huge. And one of the programs I was in didn’t actually have that at first. And then they started introducing it because they saw it like, wow, these people are coming out and they’re great and they’re great coders.
They don’t know how to tie it back to the business. Having that in mind, I think helps to how you’re going to tackle problems anyway. So those are, I mean, that’s a lot of what I look for. I like to interview. I like to just see like a combination of some of these qualities, which I understand can seem like daunting too, but it’s just, there’s just always this weight of like, we’re going to have all these great exceptional technical skills, but you’ve got to have the rest of that too.
And, you know, I think if you have Decent technical skills that you are coachable and you can expand on and you show and that you’re, you’re willing to, especially if you’re in more junior roles, then all this other stuff is like what makes you an all-star.
Evan Wimpey: That’s great. And the technical skills change as the tools update and things become more powerful.
And so that coachability, that willingness to learn, I think in a fast-moving space becomes even, even more important. For sure. I say fast moving space thinking about data and analytics specifically, but I guess in the sports betting world, like very fast moving as well. You came into the industry without any industry experience without being a sports fan.
Is that common in, in sports betting? Are you looking for people with sports or any type of betting casino type experience that come in? Or
Frazier Seay-Rad: yeah, I think it’s helpful to have passion for sure. I think a common thing is that people feel like they have to know every sport. And I’m sorry, there’s just too many sports to know everything about. Like there’s the Olympics.
Evan Wimpey: We’re recording this during the Olympics in 2024. There’s no way anybody knows all of those sports that are, I can’t follow any of it. Okay.
Frazier Seay-Rad: Sorry. And so I feel like if you, you know, just have an interest is enough and you want to know a game, right. You could know a good amount, a decent amount of basketball, right.
And like, you’re going to understand, I tell you, like, if you’ve ever seen Moneyball and you could like, kind of understand that, like, I feel like you’d be interested in this sort of field. And if any listeners haven’t watched Moneyball, I totally recommend it because it’s such a good movie. But that’s, yeah, like having an interest in it.
Like I had an interest in casino specifically. I’m. still ramping up on the sports betting side. A lot of it does make sense because at the end of the day, if you know the stock market and you kind of know about sports betting, you just may just not realize it. And it is a lot of numbers.
And one thing I know about is the stock market. So it’s very similar. A lot of it’s like interchangeable, just different lingo and terminology. Yeah. But yeah, I mean, when people, a lot of people apply, they generally have some sort of interest, whether it’s in the casino space, it’s in the sports betting space, which honestly are kind of different.
Yes, it’s all betting, but depending on like blackjack and then betting on a money line for a football game or two, I mean, So, just some sort of, some interest. A lot of people like, you know, have their favorites and they’re like, no, I’m a diehard football fan. I also just really like data science, or I really like analytics or whatever the case may be.
But you know, that’s what I ever hire somebody that necessarily didn’t have all that background. I mean, sure. Why not? You know, if you’re applying and you’re willing to learn. Yeah, absolutely. I think you can, you can learn a lot of these things pretty quickly.
Evan Wimpey: Great. I live in a state North Carolina that has, I say recently, I feel like within the last year has legalized sports.
Betting has that been a big change? I feel like sort of across the country, the legal landscape is changing on casinos and sports betting. Does that impact your work in your industry? I’m sure it impacts the industry. Does it impact the data and analytics side of your work?
Frazier Seay-Rad: Yeah, I mean, there’s a lot of preparation and getting ready for new states.
I mean, and you have to think this is a highly, highly regulated field. I mean, this is not just, I’m just going to open this app in the state because it passed, right. There’s so many hoops to jump through. So there is like a lot of thought and prep work. It takes months and months and months in advance.
I mean, you’ve got licensure requirements, regulatory requirements, and every state is a little different. So it’s not like you, you know, can have everything set up and then you’re good for the next state launch. It’s just how it works overall, because more and more states are adopting. That’s why I say this is just such a good, like a growing area.
You know, the entertainment space in itself is just—it’s just, it’s wild. Everybody’s on their phone, everybody’s on their apps, you know, that’s just—it’s just common. But yeah, I mean, there’s a ton, there’s a ton that goes on. There’s a ton for, you know, forecasting if you think about like anything with that within that space.
But yeah, there’s certainly a lot and it doesn’t slow down because the number of states that get added just in general all the time. I mean. It’s wild. So definitely not going to slow down in the, in the short run. I think when I started, I think sports betting was legal and maybe like, I don’t know, 12, 13 states or something like that. It’s just several more now, so.
Evan Wimpey: Yeah, I got to imagine that’s a very fun data science problem and forecasting problem. I think we’re working in the consulting field, but a lot of our clients that we work with have thousands or, or even tens of thousands of stores and locations. So when a new one opens.
There’s a lot of things to compare it to, but when there’s only 12 or 13 states, when it, when a new state opens, it’s a lot harder to, it’s a much bigger open than a new location two miles down the road.
Frazier Seay-Rad: Yeah. I mean, listen, every, I feel like data science in general just has such a tough time with anything new.
Right. Think like cold start problems. If you’re trying to, like, how do you recommend something to a brand-new user on your platform? How do you know about something that hasn’t really happened? You don’t have a lot of rich data and history on it. I mean, these are, these are common problems, right. And there is no like one size fits all approach to it whatsoever.
Sometimes it’s back in the napkin math and that’s sufficient. Sometimes it’s a little bit more complicated than that. But yeah, that’s typical.
Evan Wimpey: Very cool. Yeah. Simple, simple when you can make it simple, simple when simple is effective. That’s great. That sounds like an exciting problem to me. I take it.
There are numerous exciting problems. Frazier, if you, if you weren’t beholden to, to a specific schedule with priorities and you got to decide, Whatever kind of problem that you wanted to work on in data science or analytics or AI, just sort of no, no, no, not, not beholden to any other stakeholders. You just get to sort of choose what you would like to work on.
Is there anything in particular that you would think this would be a fun one? This is something I’d like to do.
Frazier Seay-Rad: Yeah, this is where I feel like I can answer this in two ways. There’s the very traditional like data science route where I think. Any kind of fraud detection. Like I’m really interested in the graph databases these days.
I think there’s so much potential. It’s newer for people that don’t really have the familiarity or the background in graph, just think finding friends of friends on a social media network and being able to start to like figure out what this information actually means. So finding meaning in very, very complex relationships and graph databases allow that I think.
That there’s a whole layer of graph data science on top of that, which is really interesting, especially in the fraud space and being able to do things very quickly. In seconds, you know, near real time. That’s like my traditional data science sort of answer. Of course I know you asked for one, I could say like personally personalizing everything, cause I think that’s just, everyone wants what they want and they want it yesterday.
I mean, that’s, you know, you expect personalization at this point. But what my ultimate answer is, it really is leaning into the AI stuff. There was a point where people were afraid of anything new and just, you know, it happens all the time. Like I think of when they thought that. The VHS tapes were going to ruin the world, and you know, it’s—the internet was going to be the end of everything.
And sure there there’s problems with everything. Right. But I think leaning into the AI space, it’s rapidly evolving. Things change every day. I get it. Sometimes it doesn’t know how many letters are in a certain word. I know it’s not great at math and that’s kind of like, not what it’s designed to do at the same time.
So but it does help me in a lot of ways. So I feel like leaning into that and figuring out what, again, keeping simplicity at the forefront is like, could you, what can you do with it? Like, can you have AI generated bet slips? Can you have AI agents? Can you have instruction, you know, there’s been studies out there that have found that people with financial advisory are AI agent about their financial situation than they are with an actual human.
So it’s like, but who would have ever thought that, right. So the more that we can get creative, I know it’s in its infancy. I just think there has to be there. There absolutely has to be so much untapped potential. And every job posting these days says AI ML, right. So again, don’t think it’s going anywhere.
I feel like there’s a lot of promising stuff. I certainly feel like data science is on the rise and analytics too. I mean, there’s everyone wants to know how something’s doing, right. Everyone wants to know the rich history of all the data that you have and how we got to where we are, right. And everyone wants to know the future.
So I don’t think that there’s really one project necessarily that I would say is above all others, except for the fact that I would lean so hard into this AI space, because it’s just, like I said, rapidly growing, helps me with A million things every day. I mean, I build study guides with it, so I’m like, and especially if you’re saying like, I want to learn let’s say it’s data science as an example, and you’re like, I don’t know.
I don’t even know where to start. I don’t blame anybody. I also didn’t know where to start. I can’t lie and say like, oh, it was an easy journey for me and here’s my recipe cookbook, right. No. Like I cried over it and it’s—that’s going to happen. It’s just that there’s a lot. But if I had the ability of what people have now to say I want to learn 80 percent of the information in 20 percent of the time, like that, give me a study guide.
I mean that’s a game changer, right. So I think these little applications could take so far. So I’m really interested to see just in general what a lot of companies start turning out with some of, some of the interesting prompts and AI use cases.
Evan Wimpey: Very cool. Very exciting. You’ve got me more and more excited about it as well. So. Frazier, thank you so much for coming on the show today. If folks want to follow what you’re up to, you’re on LinkedIn. Is that the best way?
Frazier Seay-Rad: That’s the best way.
Evan Wimpey: Okay. Follow there for data science content. No betting tips. I don’t think, although I keep waiting for them, I’m going to pounce on them when you drop one.
Frazier, thanks so much for coming on the show today. You’ve been great.
Frazier Seay-Rad: Thanks, Evan. I really appreciate it.
In this episode of the Mining Your Own Business podcast, we’re joined by Frazier Seay-Rad, a leader in AI and machine learning within the sports betting and casino industry. Frazier dives into how AI is transforming everything from enhancing customer experiences to driving revenue growth.
Evan Wimpey | Host